Safety and tolerability of fremanezumab in patients with episodic and chronic migraine: a pooled analysis of phase 3 studies

Background Fremanezumab, a fully humanized monoclonal antibody that selectively targets calcitonin gene-related peptide, has demonstrated efficacy for preventive treatment of episodic and chronic migraine. Since calcitonin gene-related peptide is expressed within the cardio- and cerebrovascular system and may have cardioprotective effects, it is critical to understand the cardio- and cerebrovascular safety of fremanezumab. Methods This was a pooled analysis of three randomized, double-blind, placebo-controlled, phase 3, 12-week trials in which patients with episodic migraine or chronic migraine received quarterly fremanezumab, monthly fremanezumab, or placebo. Incidences of overall and serious adverse events were analyzed. Cardio- and cerebrovascular adverse events (CVAEs) were analyzed in subgroups stratified by cardio- and cerebrovascular medical history, cardiovascular risk factors (CVRFs), and use of cardio- and cerebrovascular medications or triptans. Results Two thousand, eight hundred and forty-two patients were included in the study. Overall (58–65%) and serious adverse events (<1–2%) occurred in similar proportions across fremanezumab and placebo groups. CVAEs were infrequent, regardless of cardio- and cerebrovascular medical history (2–6%). CVAEs occurred in low, similar proportions of patients with CVRFs and those using cardio- and cerebrovascular medications or triptans. No cardio- and cerebrovascular signals were identified. Conclusion Fremanezumab demonstrated a favorable overall and cardio- and cerebrovascular safety profile in more than 2800 patients with episodic migraine or chronic migraine, regardless of cardio- and cerebrovascular medical history, CVRFs, or medication use. Trial Registrations: NCT02629861 (HALO EM, https://clinicaltrials.gov/ct2/show/NCT02629861), NCT02621931 (HALO CM, https://clinicaltrials.gov/ct2/show/NCT02621931), NCT03308968 (FOCUS, https://clinicaltrials.gov/ct2/ show/NCT03308968)


Introduction
Migraine is a complex and burdensome disease that affects >1 billion individuals and is the second leading cause of years lived with disability worldwide (1). Episodic migraine (EM), and especially chronic migraine (CM), have a substantial negative effect on health-related quality of life for patients (2). Preventive medications are recommended in patients with frequent or disabling attacks to reduce the frequency of migraine attacks and/or to mitigate their impact. While there are numerous options available for migraine preventive therapy, such as betablockers, antihypertensives, anticonvulsants, antidepressants, and onabotulinumtoxinA, these medications were not developed specifically for migraine (3). Among these therapies, only onabotulinumtoxinA is approved for the preventive treatment of CM (4). Due in part to suboptimal tolerability, efficacy, and adherence to these medications, there was an unmet need for migraine-specific preventive treatments with proven tolerability, safety, and efficacy in the reduction of both frequency, severity, and duration of migraine attacks (5).
The development of monoclonal antibodies (mAbs) that target the calcitonin gene-related peptide (CGRP) pathway has significantly advanced migraine preventive therapy. Four CGRP pathway-targeting mAbs (fremanezumab, erenumab, galcanezumab, and eptinezumab) are approved for the preventive treatment of EM and CM (6). Fremanezumab, galcanezumab, and eptinezumab target the CGRP ligand, while erenumab targets the CGRP receptor. CGRP is a neuropeptide that is produced in peripheral sensory neurons and the central nervous system and is integral to the pathophysiology of migraine (7). Blockade of the CGRP pathway has not only been found to be effective in migraine prevention, but also in aborting migraine attacks (6).
In addition to the pial and dural vasculature, CGRP receptors are expressed within the cardio-and cerebrovascular (CV) system and are localized in sensory nerve fibers and around the peripheral arteries of the heart (8,9). CGRP is a potent vasodilator that has a role in arterial function and has been hypothesized to regulate local vascular responses within the CV system, including mediating collateral recruitment during cerebral ischemia (8,10). Because of the potential cardioprotective capacity of CGRP, understanding the CV safety of its inhibition by CGRP pathway-targeting mAbs is of critical importance (8).
Triptans, a first-line option for the acute treatment of migraine attacks, are selective 5-hydroxytriptamine (5-HT 1B/1D ) receptor agonists. 5-HT 1B/1D receptors are expressed in CV and coronary blood vessels, and the stimulation of the 5-HT 1B receptor results in minor cranial vasoconstriction (11). Because of their vasoconstrictive capacity, triptans can theoretically increase the risk of serious ischemic events and have therefore been contraindicated in patients with significant CV disease (11)(12)(13). However, this recommendation was based on the incorrect assumption that most acute coronary syndromes and ischemic strokes are caused by vasoconstriction, and a study by Goldstein et al. showed that even high doses of triptans did not significantly constrict coronary artery diameter (14). Furthermore, a retrospective cohort study of 130,411 patients with migraine and 130,411 matched controls showed no increase in the risk of myocardial infarction with current or recent triptan use, supporting the CV safety of triptans (15). Nevertheless, given that triptans are used for acute migraine treatment alongside preventive medications, including CGRP pathway-targeting mAbs, and that each of these drugs has a potential role in CV function, it is important to confirm the CV safety of their concomitant use.
Fremanezumab is a fully humanized mAb (IgG2Da) that selectively targets CGRP and is approved for the preventive treatment of migraine in adults (16). It has demonstrated good safety, tolerability, and efficacy for the preventive treatment of migraine in patients with EM or CM, including those with difficult-to-treat migraine, in three randomized, double-blind, placebocontrolled, phase 3 trials (17)(18)(19). The long-term safety and efficacy of fremanezumab treatment have been demonstrated for up to 12 months in patients with EM or CM in a randomized, double-blind, parallelgroup, phase 3 study (20). Here, we present the overall and CV safety of fremanezumab over 12 weeks of the placebo-controlled period from three phase 3 trials in patients with EM or CM with or without CV medical history, with CV risk factors (CVRFs), and with concomitant use of medications for the treatment of CV disease or the intake of triptans.  (17)(18)(19). Eligible participants for all the studies were 18 to 70 years of age, with a history of migraine with onset at or prior to age 50 years and for at least 12 months prior to screening. The HALO EM study included patients with EM (6 and <15 headache days per month, with 4 days fulfilling the International Classification of Headache Disorders 3 beta version [ICHD-3 beta] criteria (21) for migraine with aura or without aura, probable migraine, or use of triptans or ergots to treat an established headache). The HALO CM study included patients with CM (15 headache days per month, with 8 days fulfilling the ICHD-3 beta criteria for migraine with or without aura, probable migraine, or use of triptans or ergots to treat an established headache). The FOCUS study included patients with EM or CM and documented inadequate response (no clinically meaningful improvement per treating physician's judgment after 3 months of therapy; treatment had to be interrupted because of adverse events (AEs) that made it intolerable for the patients, or the medication was contraindicated or unsuitable for the preventive treatment of migraine for the patient) to 2 to 4 classes of migraine preventive medications (betablockers, tricyclic antidepressants, calcium channel blockers, angiotensin II receptor antagonists, onabotulinumtoxinA, or valproic acid) within the past 10 years (19).

Ethics approvals and patient consent
The studies included in this analysis were conducted in compliance with the International Conference for Harmonisation guidelines for Good Clinical Practice, the Declaration of Helsinki, and relevant national and local regulations. Patients from each study provided written informed consent before study procedures and assessments were conducted.

Randomization and treatment procedures
Patients were randomly assigned 1:1:1 to quarterly fremanezumab (675 mg fremanezumab at baseline, matched monthly placebo at weeks 4 and 8), monthly fremanezumab (for patients with EM: 225 mg fremanezumab at baseline, week 4, and week 8; for patients with CM: 675 mg fremanezumab at baseline, 225 mg fremanezumab at weeks 4 and 8), or matched monthly placebo by subcutaneous (SC) injection.
Randomization was performed using electronic interactive response technology, and patients were stratified according to baseline preventive medication use, sex, country, and migraine classification (for the FOCUS study). The sponsor, patients, investigators, and designated personnel were blinded to treatment assignments.

Safety endpoints
Safety and tolerability endpoints included incidence of overall AEs, serious AEs, and AEs leading to discontinuation. The incidence of all cardio-or cerebrovascular AEs (CVAEs), including arrhythmias and electrocardiogram changes, was assessed in several patient subgroups: patients with and without CV medical history (i.e. CV disease-related medical conditions reported in the patient's medical history); patients with CVRFs (i.e. potential risk factors for developing CV disease); patients who were receiving CV medication(s) at baseline; and patients who used triptans throughout the study. While patients with a history of serious CV disease, such as myocardial infarction or stroke, were excluded from this study, patients with CV medical history (Table S1) and CVRFs (Table S2) were included.

Data analysis
The patient population presented here comprises the safety populations from each study, which consisted of all patients who received 1 dose of the study drug. Descriptive statistics are reported for continuous and categorical variables. For continuous variables, mean and standard deviation (SD) are reported. For categorical variables, count (n) and percentage of patients in each category are provided. No statistical analyses are reported in this study.

Patient population
A total of 2842 patients were included in the pooled safety population. Demographic characteristics were similar across treatment groups ( Table 2). Between 85% and 86% of patients in each treatment group were female, and the proportion of patients with EM ranged from 42% to 43% ( Table 2). The most prevalent CV medical history and CVRFs among the study population are provided in Tables S1 and S2, respectively. The most commonly reported CV medical history was hypertension (quarterly fremanezumab, 8%; monthly fremanezumab, 8%; total fremanezumab, 8%; placebo, 9%; Table S1). Other CV medical history was reported in 1% of patients across treatment groups (Table S1). The most common CVRFs were obesity (22-25%), history of CV disease (15-16%), and hormonal birth control pill use (18-19%; Table S2).

Overall safety and tolerability
The proportion of patients who experienced 1 AE was similar across treatment groups (quarterly fremanezumab, 65%; monthly fremanezumab, 62%; total fremanezumab, 63%; placebo, 58%; Table 3). Injection-site pain, injection-site induration, and injection-site erythema were the most commonly reported AEs and had similar incidences across treatment groups (Table 3). Apart from nausea (quarterly fremanezumab, 2%; monthly fremanezumab, 1%; total fremanezumab, 1%; placebo, 2%), gastrointestinal AEs were reported in <2% of patients in each of the fremanezumab treatment groups. Serious AEs and AEs leading to discontinuation from the study drug occurred in 2% of patients across treatment groups (Table S3). The most commonly reported AEs leading to discontinuation were also injection site related (erythema and rash); each of these led to discontinuation at low, similar rates across treatment groups (all groups, <1%). Two deaths occurred during the trials; one patient in HALO EM died by intentional overdose of diphenhydramine 109 days after receiving a single quarterly dose of fremanezumab and 38 days after withdrawing consent from the study, and one patient in HALO CM died of chronic obstructive pulmonary disease 69 days after receiving a single quarterly dose of fremanezumab. An autopsy was performed in both cases to confirm the cause of death; intake of the study drug was not considered causal for either death (17,18).

CV safety in patients with CV medical history or risk factors
To evaluate the CV safety of fremanezumab with respect to relevant patient characteristics, we analyzed the incidence of CVAEs in patient subgroups according to CV medical history, CVRFs, and concomitant CV medication and triptan use. Among patients with CV medical history, the incidence of CVAEs ranged from 3% to 6% (Table 4). Hypertension was the most commonly reported CVAE among patients with CV medical history but did not occur in >2% of patients with CV medical history in any treatment group (quarterly fremanezumab, 2%; monthly fremanezumab, <1%; total fremanezumab, 1%; placebo, <1%; Table S4). No other CVAE occurred in >1% of patients with CV medical history. CVAEs occurred in low, similar proportions of patients without CV medical history (all treatment groups, 2%; Table S4). Similar to patients with CV medical history, hypertension, while rare, was the most frequently reported CVAE among patients without CV medical history (<1%, all treatment groups; Table S4).
CVAEs occurred at similar, low rates across treatment groups in patients with 2 CVRFs without CV medical history (quarterly fremanezumab, 2%; monthly fremanezumab, 4%; total fremanezumab, 3%; placebo, 4%; Table S5), and none of the CVAEs reported occurred in >1 patient in any individual treatment group. In patients with 4 CRVFs, no CVAEs were reported in any treatment group. In a 12-month follow-up study of long-term safety, tolerability, and efficacy of fremanezumab that

CV safety in patients with baseline and concomitant medication use
The incidence of AEs was similar among patients who used CV medications at baseline compared to the total patient population (quarterly fremanezumab, 66%; monthly fremanezumab, 64%; total fremanezumab, 65%; placebo, 53%; Table 6). Similar to the total patient population, the most commonly reported AEs among patients in this subgroup were injection site related (pain, erythema, induration). Injection siterelated AEs, most notably injection-site induration, occurred more frequently in the fremanezumab groups compared to the placebo group in patients receiving CV medications at baseline (injection-site induration: quarterly fremanezumab, 18%; monthly fremanezumab, 24%; total fremanezumab, 21%; placebo, 8%; Table 6). CVAEs occurred at low, similar rates across treatment groups in patients receiving CV medications at baseline, and none of the CVAEs reported occurred in >1 patient (Table 6).
In patients with concomitant triptan use, CVAEs occurred in low, similar proportions across treatment groups (all groups, 2%; Table 7). Increased heart rate and hypertension were the most commonly reported CVAEs in this subgroup; each occurred in <1% of patients in each treatment group. CVAEs were also infrequent in patients without concomitant triptan use (all groups, 3%), and each CVAE occurred in 1% of patients (Table S6).

Discussion
This pooled analysis showed that fremanezumab was safe and well tolerated in a cohort of >2800 patients with migraine from three randomized, double-blind, placebo-controlled, phase 3 clinical trials. The studies included patients with EM and CM, as well as patients with previous inadequate response to two to four classes of migraine preventive medications, thus demonstrating safety and tolerability even in patients with difficult-to-treat migraine. Among patients with migraine, AEs such as weight gain, memory loss, and depression, as well as lack of efficacy, are the most common reasons for discontinuation of migraine preventive treatment (5,22). One longitudinal study found that 75% of patients with CM using oral migraine preventive treatments discontinued treatment by 6 months, and an additional 10% of patients discontinued treatment by 12 months (23). In another study, 24% of patients with EM reported discontinuation of 1 preventive medication (24). Compliance for onabotulinumtoxinA is also low; a year-long study reported that only 12.6% of patients followed were treated for 12 months with the prescribed number of cycles (25). The safety profile shown for fremanezumab in a large patient population in this analysis, along with the efficacy data from phase 3 clinical trials (17)(18)(19)(20), support its potential as an option for patients seeking safe, tolerable, and efficacious migraine-specific preventive medication.
Fremanezumab is one of four CGRP pathwaytargeting mAbs approved for the preventive treatment of migraine, which represent a significant advancement in migraine-specific preventive therapy.  Galcanezumab, erenumab, and eptinezumab have generally favorable safety profiles, with serious AEs and AEs leading to discontinuation occurring infrequently across treatment groups (6). The long-term safety of fremanezumab, galcanezumab, and erenumab has also been reported over 12 months of treatment (20,26,27). Fremanezumab, galcanezumab, and erenumab are each administered by SC injection, while eptinezumab is administered by intravenous infusion. Injection-site reactions are commonly reported AEs for those CGRP pathway-targeting antibodies administered by SC injection. Injection-site reactions have been cited as reasons for discontinuation in trials of fremanezumab and galcanezumab, albeit infrequently relative to the occurrence of injection-site reactions (28). In addition to overall safety and tolerability for patients with migraine, the CV safety of CGRP pathway-targeting antibodies has been a central concern due to CGRP's role in the cardiovascular system. Here, we show that treatment with fremanezumab over 12 weeks has a favorable CV safety profile, even in patients with CV medical history or CVRFs. Fremanezumab was also safe and tolerable in patients receiving CV medications or triptans, and no safety signals were identified across the three studies included  in this analysis. The long-term CV safety of fremanezumab has been demonstrated in a 12-month, doubleblind, parallel-group study of >1800 patients, in which CVAEs were infrequent and mostly mild to moderate in severity (20). In that study, similar to the safety profile reported here, hypertension was the most commonly reported CVAE. Hypertension only occurred in 2% of patients, and the majority of those events were single blood pressure elevations in patients with a history of hypertension and were assessed as not related to the study drug (20). For patients with hypertension, patients with hypertension at baseline, and patients using concomitant antihypertensive medications, blood pressure values were similar or numerically lower at the end of up to 15 months of treatment than at baseline, suggesting that fremanezumab did not have an adverse impact on blood pressure in those patients. Possible explanations for the numerically lower post-treatment blood pressure values in patients with hypertension include a secondary effect of having fewer migraine days or improved compliance with antihypertensive medications, although this reasoning is speculative and has not been tested. Likewise, all serious CVAEs in the long-term study were considered not related to the study drug (20). These studies demonstrate the short-and long-term CV safety of fremanezumab for patients with migraine, including those with CV clinical features.  CV safety has also been investigated for galcanezumab and erenumab. In a pooled analysis of six phase 2 and 3 trials of erenumab in patients with EM or CM, vascular AEs were infrequent and occurred in similar proportions of patients across treatment groups (29). There was no association between erenumab treatment and vascular events, including in patients with vascular risk factors at baseline (29). Subgroup analyses of patients using migraine-specific medications showed that the incidences of AEs were similar in patients receiving erenumab or placebo, regardless of the use of triptans or ergots (29). Similar to fremanezumab, hypertension was the most commonly reported CVAE in the erenumab trials; however, its incidence was infrequent and similar between erenumab and placebo treatment groups (29).
In the post-marketing setting, a number of reports have detailed the development of hypertension and worsening of pre-existing hypertension following the use of erenumab, leading to an update to the US prescribing information to include a warning regarding this risk (30). Many of the patients had pre-existing hypertension or risk factors for hypertension. There were cases requiring pharmacologic treatment and, in some cases, hospitalization. Hypertension occurred at different times during treatment but was most frequently reported within 7 days of dose administration. In the majority of the cases, the onset or worsening of hypertension was reported after the first dose, and erenumab was discontinued in many of the reported cases (30). A definitive cause for this observation has not been identified, although it is possible that the mechanism of action for erenumab, which involves targeting the CGRP receptor and is unique among the approved CGRP pathway-targeting mAbs, may be a contributing factor. To date, development or worsening of hypertension has not been reported in the postmarketing setting for fremanezumab, galcanezumab, or eptinezumab (16,31,32).
A pooled analysis of three phase 3 trials of up to 6 months of galcanezumab treatment in patients with EM or CM demonstrated a favorable CV safety profile. Galcanezumab treatment was not associated with an increased incidence of CVAEs compared to placebo, nor changes in blood pressure, pulse, or electrocardiogram features compared to placebo (33). The incidence of hypertension was similar between placebo and galcanezumab treatment groups. A subgroup analysis of patients with concomitant triptan use did not reveal increased risk for CVAEs (33). Together, these studies support the overall and CV safety of CGRP pathwaytargeting mAbs, even in individuals who may be at an increased risk for adverse CV outcomes due to CV medical history, CVRFs, or CV medication or triptan use. Although these results are encouraging for the treatment outlook for patients with CV clinical features and migraine, patients at high risk for myocardial infarction or ischemic stroke were excluded from this study and caution should be exercised when prescribing mAbs due to the risk of hypersensitivity-associated acute coronary syndromes (34). A long-term observational safety study of fremanezumab with a nested cohort of CV-compromised patients is currently being conducted to further evaluate the real-world overall safety and CV safety of treatment.

Conclusion
This pooled safety study showed that fremanezumab treatment was generally safe and well tolerated over 12 weeks in patients with EM or CM, including those with difficult-to-treat migraine. Fremanezumab treatment also showed a favorable CV profile, even in individuals with CV medical history, CVRFs, or concomitant CV medication or triptan use. Patients with high risk of myocardial infarction or stroke were excluded from these studies. Our data support fremanezumab as reasonable for preventive migraine therapy in patients with EM or CM regardless of their CV clinical features.

Key findings
• Fremanezumab was safe and well tolerated in a pooled cohort of >2800 patients with episodic or chronic migraine, including those with difficult-to-treat migraine. • Fremanezumab has a favorable cardio-/cerebrovascular (CV) safety profile, even in patients with CV medical history and/or CV risk factors. • Fremanezumab is safe and well tolerated in patients with concomitant use of cardiovascular medications or triptans. PhD, of Cello Health Communications/MedErgy. This assistance was in accordance with Good Publication Practice (GPP3) guidelines and funded by Teva Pharmaceuticals. The authors maintained full editorial control of the manuscript and the decision to submit for publication.

Declaration of conflicting interests
The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: HCD received honoraria for participation in clinical trials, contribution to advisory boards, or oral presentations from: Alder, Allergan, Amgen, electroCore, Ipsen, Lilly, Medtronic, Novartis, Pfizer, Teva Pharmaceuticals, and Weber & Weber. electroCore provided financial support for research projects. The German Research Council, the German Ministry of Education and Research, and the European Union support his headache research. PM is an employee of the New England Institute for Neurology and Headache. He is an independent contractor and part of speaker bureaus at Teva Pharmaceuticals, Amgen, and Allergan. TPJ received honoraria for contribution to advisory boards or oral presentations from: Allergan, Desitin, Hormosan, Novartis, Lilly, Sanofi, and Teva Pharmaceuticals. His research is supported by Gemeinsame Bundesausschuss -Innovationsfonds and EFRE. YK, XN, VRC, and SB are employees and/or stockholders of Teva Branded Pharmaceutical Product R&D, Inc. JMC is a former employee of Teva Branded Pharmaceutical Product R&D, Inc. SDS provides consultation to Alder, Allergan, Amgen, Autonomic Technologies, Avanir, Curelater, Inc., Depomed, Dr. Reddy's Laboratories, Ensured, Inc., electroCore, eNeuraTherapeutics, INSYS Therapeutics, Lilly, Supernus Pharmaceuticals, Inc., Teva Pharmaceuticals, Theranica, and Trigemina, Inc.